Compare Q>=3 and Q>=4 Datasets: Participant Characteristics

Participant Table, Q>=4

## 
## 
## |                      | level  |          ASD          |           TD           |   p    | test |
## |:-------------------------:|:------:|:---------------------:|:----------------------:|:------:|:----:|
## |           **n**           |        |          28           |           27           |        |      |
## |        **SEX (%)**        | Female |       6 (21.4)        |       14 (51.9)        | 0.039  |      |
## |                           |  Male  |       22 (78.6)       |       13 (48.1)        |        |      |
## |  **V1.Age (mean (SD))**   |        |     39.39 (13.77)     |     39.56 (13.98)      | 0.965  |      |
## |  **MEL_cat (mean (SD))**  |        |      3.62 (1.55)      |      4.41 (1.05)       | 0.033  |      |
## |    **INR (mean (SD))**    |        |      3.33 (2.54)      |      4.43 (2.09)       | 0.087  |      |
## | **zipIncome (mean (SD))** |        |  53477.39 (18001.70)  |  57480.76 (15744.99)   | 0.439  |      |
## |   **SES1 (mean (SD))**    |        |     -0.40 (2.06)      |      0.62 (1.76)       | 0.054  |      |
## | **ExpLang_T (mean (SD))** |        |     31.67 (11.07)     |     49.78 (11.10)      | <0.001 |      |
## | **RecLang_T (mean (SD))** |        |     32.32 (11.85)     |     53.85 (10.96)      | <0.001 |      |
## |  **ELC_SS (mean (SD))**   |        |     71.96 (20.00)     |     105.67 (15.78)     | <0.001 |      |
## |    **TBV (mean (SD))**    |        | 1077031.79 (96937.48) | 1050164.41 (111924.89) | 0.345  |      |
## 
## Table: Participant Summary Table for Q>=4

Participant Table, Q>=3

## 
## 
## |          &nbsp;           | level  |          ASD           |           TD           |   p    | test |
## |:-------------------------:|:------:|:----------------------:|:----------------------:|:------:|:----:|
## |           **n**           |        |           36           |           31           |        |      |
## |        **SEX (%)**        | Female |       10 (27.8)        |       14 (45.2)        | 0.221  |      |
## |                           |  Male  |       26 (72.2)        |       17 (54.8)        |        |      |
## |  **V1.Age (mean (SD))**   |        |     39.03 (12.95)      |     37.61 (14.20)      | 0.671  |      |
## |  **MEL_cat (mean (SD))**  |        |      3.61 (1.48)       |      4.42 (0.99)       | 0.013  |      |
## |    **INR (mean (SD))**    |        |      3.36 (2.53)       |      4.22 (2.14)       | 0.149  |      |
## | **zipIncome (mean (SD))** |        |  54158.10 (18734.17)   |  56118.68 (15004.41)   | 0.675  |      |
## |   **SES1 (mean (SD))**    |        |      -0.35 (1.97)      |      0.54 (1.66)       | 0.053  |      |
## | **ExpLang_T (mean (SD))** |        |     32.62 (11.76)      |     48.77 (11.52)      | <0.001 |      |
## | **RecLang_T (mean (SD))** |        |     32.49 (12.52)      |     53.32 (11.13)      | <0.001 |      |
## |  **ELC_SS (mean (SD))**   |        |     72.23 (19.97)      |     104.94 (16.57)     | <0.001 |      |
## |    **TBV (mean (SD))**    |        | 1076827.78 (100384.08) | 1045805.34 (115665.34) | 0.244  |      |
## 
## Table: Participant Summary Table for Q>=3

LGI Results, Bivariate Correlations

Bivariate Correlation Plots, ASD&TD Together, Q>=4

Bivariate Plot

Bivariate Plot

Bivariate Correlation ASD and TD, Q>=3

Bivariate Plot

Bivariate Plot

Bivariate correlations, ASD Only, Q>=4

Bivariate Plot ASD Only

Bivariate Plot ASD Only

Bivariate correlations, ASD Only, Q>=3

Bivariate Plot ASD Only

Bivariate Plot ASD Only

Bivariate correlations, TD Only Q>=4

Bivariate Plot, TD Only

Bivariate Plot, TD Only

Bivariate correlations, TD Only Q>=3

Bivariate Plot, TD Only

Bivariate Plot, TD Only

LGI Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models: INR

INR Results Q>=4

  rh_parsopercularis_lgi
Coeffcient Estimates CI (95%) p-Value
Intercept 3.49 2.58 – 4.40 <0.001
Income:Needs 0.03 -0.01 – 0.06 0.104
TBV 0.00 -0.00 – 0.00 0.055
Age 0.01 -0.00 – 0.01 0.157
Observations 53
R2 / R2 adjusted 0.236 / 0.189

  • Income to needs ratio predicts rh_parsopercularis_lgi, controlling for age and TBV in the Q>=4 dataset.

INR Results Q>3

  rh_parsopercularis_lgi
Coeffcient Estimates CI (95%) p-Value
Intercept 3.64 2.86 – 4.42 <0.001
Income:Needs 0.03 -0.01 – 0.06 0.104
TBV 0.00 0.00 – 0.00 0.039
Age 0.00 -0.00 – 0.01 0.245
Observations 64
R2 / R2 adjusted 0.202 / 0.162

  • Income to needs ratio predicts rh_parsopercularis_lgi, controlling for age and TBV in the Q>=3 dataset.

LGI Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models SES1

Neighborhood Advantage Results Q>=4

  • Neighborhood Advantage not a significant predictor of LGI when controlling for TBV and dx for Q>=4.

Neighborhood Advantage Results Q>3

  • Neighborhood Advantage not a significant predictor of LGI when controlling for TBV and dx for data Q>=3.

LGI Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models: MEL

MEL Results Q>=4

  • Maternal Education not a significant predictor of LGI when controlling for TBV and dx for data Q>=4.

MEL Results Q>=3

  • Maternal Education not a significant predictor of LGI when controlling for TBV and dx for data Q>=3.

LGI Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models ZipIncome

Zip-Income Results, data Q>=4

  • Zip-Income is a significant predictor or lh pars orbitalis LGI when controlling for age (removed TBV because it was not a significant predictor in the model with age and Zip-Income).

  • Dx not a significant predictor in this model, nor is the diagnosisXzipIncome interaction term.

  lh_parsorbitalis_lgi
Coeffcient Estimates CI (95%) p-Value
Intercept 2.68 2.36 – 3.01 <0.001
Zip-Income 0.00 0.00 – 0.00 0.018
Age 0.01 0.00 – 0.01 0.019
Observations 44
R2 / R2 adjusted 0.235 / 0.198

Zip-Income Results Q>=3

  • Zip-Income a significant predictor of lh-pars orbitals LGI, when controlling for age (removed TBV and dx bc they were not significant predictors in this model), for Q>=3.
  lh_parsorbitalis_lgi
Coeffcient Estimates CI (95%) p-Value
(Intercept) 2.66 2.38 – 2.94 <0.001
zipIncome 0.00 0.00 – 0.00 0.002
V1.Age 0.01 0.00 – 0.01 0.014
Observations 55
R2 / R2 adjusted 0.267 / 0.239

LGI Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models: Dx

Dx Results, Q>=4

  • Dx predicts left hemisphere pars triangularis and right hemisphere middle temporal LGI, when controlling for TBV, age, and sex (the latter were not sig. predictors for RH MTG LGI, so removed from the model).

  • Participants with ASD show higher lh pars triangularis LGI than TD participants, controlling for other variables.

  • ASD participants show lower rh MTG LGI than TD participants.

  lh_parstriangularis_lgi
Coeffcient Estimates CI (95%) p-Value
Intercept 3.26 2.51 – 4.01 <0.001
Dx -0.15 -0.29 – -0.01 0.034
TBV 0.00 0.00 – 0.00 0.001
Sex -0.18 -0.35 – -0.01 0.038
Observations 55
R2 / R2 adjusted 0.234 / 0.189
  rh_middletemporal_lgi
Coeffcient Estimates CI (95%) p-Value
Intercept 2.29 1.83 – 2.76 <0.001
Dx 0.09 0.01 – 0.18 0.035
TBV 0.00 0.00 – 0.00 <0.001
Observations 54
R2 / R2 adjusted 0.393 / 0.369

Dx Results, Q>=3

  • Diagnosis not significantly associated with LGI in the Q>=3 dataset (after removing SES variables from the model). Including results here for comparison to the Q4 model results.
  lh_parstriangularis_lgi
Coeffcient Estimates CI (95%) p-Value
Intercept 3.01 2.33 – 3.69 <0.001
Dx -0.10 -0.23 – 0.03 0.116
TBV 0.00 0.00 – 0.00 <0.001
Sex -0.14 -0.28 – 0.01 0.059
Observations 67
R2 / R2 adjusted 0.262 / 0.227
  rh_middletemporal_lgi
Coeffcient Estimates CI (95%) p-Value
Intercept 2.53 2.10 – 2.97 <0.001
Dx 0.05 -0.03 – 0.13 0.235
TBV 0.00 0.00 – 0.00 <0.001
Observations 66
R2 / R2 adjusted 0.299 / 0.277

LGI Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models: Sex/Gender

Sex/Gender Results, Q>=4

  • Sex predicts lh pars opercularis LGI, controlling for TBV, dx (removed age bc it was not significant). Boys show higher lh pars opercularis LGI than girls.
  lh_parsopercularis_lgi
Coeffcient Estimates CI (95%) p-Value
Intercept 3.12 2.32 – 3.93 <0.001
Sex -0.22 -0.40 – -0.04 0.020
Dx -0.14 -0.29 – 0.01 0.065
TBV 0.00 0.00 – 0.00 <0.001
Observations 55
R2 / R2 adjusted 0.359 / 0.321

Sex Results Q>=3

  • Sex predicts lh pars opercularis LGI, controlling for TBV and dx (removed age bc it was insignificant). Boys show higher lh pars opercularis LGI than girls.
  lh_parsopercularis_lgi
Coeffcient Estimates CI (95%) p-Value
Intercept 3.15 2.47 – 3.82 <0.001
Sex -0.19 -0.33 – -0.05 0.009
Dx -0.14 -0.27 – -0.01 0.031
TBV 0.00 0.00 – 0.00 <0.001
Observations 67
R2 / R2 adjusted 0.423 / 0.395

LGI Results, Associations with Behavioral Data

Bivariate correlations with bx data, ASD Only, Q>=4

Bivariate Plot ASD Only, with Bx Data

Bivariate Plot ASD Only, with Bx Data

Bivariate correlations with bx data, ASD Only, Q>=3

Bivariate Plot ASD Only, with Bx Data

Bivariate Plot ASD Only, with Bx Data

Bivariate correlations with bx data, TD Only, Q>=4

Bivariate Plot TD Only, with Bx Data

Bivariate Plot TD Only, with Bx Data

Bivariate correlations with bx data, TD Only, Q>=3

Bivariate Plot TD Only, with Bx Data

Bivariate Plot TD Only, with Bx Data

Regression Results for Bx Models, Q>=4

  • Receptive Language is postively associated with rh pars orbitalis LGI, controlling for dx and age for Q>=4 (TBV not a significant predictor, removed from the model). I should probably make partial regression plots to better visualize the assocations.

  • Insignificant negative relationship between receptive language and lh pars orbitalis LGI that is significant in the Q>=3 dataset (see below). The negative effect looks to be mostly driven by the ASD group, although there is no significant RecLangXGroup interaction.

  • Both expressive and receptive language predict rh and lh fusiform LGI, but not including in this summary as it's not directly related to the dissertation, but noting for later. ELC also associated with fusiform LGI.

  rh_parsopercularis_lgi
Coeffcient Estimates CI (95%) p-Value
Intercept 4.45 4.06 – 4.84 <0.001
Rec Lang 0.00 -0.00 – 0.01 0.326
Dx -0.12 -0.37 – 0.12 0.317
Age 0.01 0.00 – 0.02 0.006
Observations 54
R2 / R2 adjusted 0.145 / 0.094
  lh_parsopercularis_lgi
Coeffcient Estimates CI (95%) p-Value
Intercept 4.44 4.09 – 4.79 <0.001
Rec Lang 0.01 -0.00 – 0.01 0.157
Dx -0.24 -0.46 – -0.02 0.030
Age 0.01 0.00 – 0.02 <0.001
Observations 55
R2 / R2 adjusted 0.253 / 0.209
  rh_parsopercularis_lgi
Coeffcient Estimates CI (95%) p-Value
Intercept 4.75 4.44 – 5.06 <0.001
Exp Lang 0.01 -0.00 – 0.02 0.101
Dx -0.17 -0.42 – 0.08 0.173
Observations 53
R2 / R2 adjusted 0.055 / 0.017

  • Receptive Language is not significantly associated with rh pars orbitalis LGI for the Q>=3 dataset, in contrast to the Q>=4 results, controlling for dx and age (TBV not a significant predictor, removed from the model).

  • Significant, negative relationship between receptive language and lh pars orbitalis LGI in the Q>=3 dataset (was not quite significant in the Q>=4 dataset) for ASD group, but not for TD group. The RecLangXGroup interaction is significant in this dataset.

  • Can run a special regression model designed for truncated data distributions (e.g., floor effects) in order to verify that this result is not driven just by violations of homoscedasticity assumtion (if we decide to stick with this dataset. . .).

  • No significant relationship between expressive language and rh pars opercularis LGI, controlling for dx, TBV, and age in the Q>=3 dataset (in contrast to Q>=4).

Regression Results for Bx Models, Q>=3

  rh_parsopercularis_lgi
Coeffcient Estimates CI (95%) p-Value
Intercept 4.50 4.16 – 4.85 <0.001
Rec Lang 0.00 -0.00 – 0.01 0.434
Dx -0.08 -0.28 – 0.13 0.464
Age 0.01 0.00 – 0.01 0.006
Observations 65
R2 / R2 adjusted 0.122 / 0.079
  lh_parsorbitalis_lgi
Coeffcient Estimates CI (95%) p-Value
Intercept 3.91 3.27 – 4.54 <0.001
Rec Lang -0.02 -0.04 – -0.01 0.005
Dx -0.53 -1.02 – -0.04 0.034
Age 0.01 0.00 – 0.01 0.017
RL by Dx Intxn 0.01 0.00 – 0.02 0.019
Observations 66
R2 / R2 adjusted 0.215 / 0.164
  rh_parsopercularis_lgi
Coeffcient Estimates CI (95%) p-Value
Intercept 4.78 4.50 – 5.06 <0.001
Exp Lang 0.00 -0.00 – 0.01 0.237
Dx -0.10 -0.31 – 0.10 0.327
Observations 64
R2 / R2 adjusted 0.025 / -0.007

CT Results, Bivariate Correlations

Bivariate Correlation Plots, ASD&TD Together, Q>=4

Bivariate Plot

Bivariate Plot

Bivariate Correlation ASD and TD, Q>=3

Bivariate Plot

Bivariate Plot

Bivariate correlations, ASD Only, Q>=4

Bivariate Plot ASD Only

Bivariate Plot ASD Only

Bivariate correlations, ASD Only, Q>=3

Bivariate Plot ASD Only

Bivariate Plot ASD Only

Bivariate correlations, TD Only Q>=4

Bivariate Plot, TD Only

Bivariate Plot, TD Only

Bivariate correlations, TD Only Q>=3

Bivariate Plot, TD Only

Bivariate Plot, TD Only

CT Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models: INR

INR Results Q>=4

  rh_superiortemporal_CT
Coeffcient Estimates CI (95%) p-Value
Intercept 3.36 3.16 – 3.56 <0.001
Income:Needs 0.02 -0.00 – 0.04 0.057
Age -0.00 -0.01 – -0.00 0.003
Sex 0.04 -0.05 – 0.13 0.345
Observations 54
R2 / R2 adjusted 0.222 / 0.176

  • INR Predicts LH middletemporal cortical thickness (CT), controlling for age and sex/gender in the Q>=4 dataset.

INR Results Q>3

  rh_parstriangularis_CT
Coeffcient Estimates CI (95%) p-Value
Intercept 2.78 2.59 – 2.97 <0.001
Income:Needs 0.01 -0.00 – 0.03 0.114
Age -0.00 -0.01 – -0.00 0.042
Sex 0.09 0.01 – 0.18 0.032
Observations 68
R2 / R2 adjusted 0.141 / 0.100
  lh_middletemporal_CT
Coeffcient Estimates CI (95%) p-Value
Intercept 3.32 3.14 – 3.49 <0.001
Income:Needs 0.02 0.00 – 0.04 0.019
Age -0.00 -0.01 – -0.00 0.003
Sex 0.03 -0.04 – 0.11 0.393
Observations 68
R2 / R2 adjusted 0.190 / 0.152
  lh_superiortemporal_CT
Coeffcient Estimates CI (95%) p-Value
Intercept 3.04 2.86 – 3.21 <0.001
Income:Needs 0.01 -0.00 – 0.03 0.074
Age -0.00 -0.00 – 0.00 0.359
Sex 0.01 -0.07 – 0.09 0.816
Observations 68
R2 / R2 adjusted 0.058 / 0.014

  • INR Predicts RH pars triangularis, LH middletemporal, and LH superiortemporal cortical thickness (CT), controlling for age and sex/gender in the Q>=3 dataset.

CT Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models: SES1

SES1 Results Q>=4

  • SES1 predicts rh middle temporal CT (controlling for age and sex).

  • SES1 predicts lh pars opercularis CT (controlling for age and sex).

  • TBV not a significant predictor of CT generally.

  lh_parsopercularis_CT
Coeffcient Estimates CI (95%) p-Value
Intercept 3.13 2.98 – 3.29 <0.001
SES1 0.02 0.00 – 0.04 0.044
Age -0.01 -0.01 – -0.00 <0.001
Sex 0.03 -0.05 – 0.10 0.456
Observations 55
R2 / R2 adjusted 0.284 / 0.242
  rh_middletemporal_CT
Coeffcient Estimates CI (95%) p-Value
Intercept 3.25 3.07 – 3.42 <0.001
SES1 0.02 0.00 – 0.04 0.039
Age -0.00 -0.01 – -0.00 0.017
Sex 0.08 -0.00 – 0.16 0.061
Observations 55
R2 / R2 adjusted 0.196 / 0.148

SES1 Results Q>3

  • Associations between SES1 and pars opercularis / middle temporal gyrus are not significant in the Q>=3 dataset, controlling for age, sex/gender, and TBV.
  rh_middletemporal_CT
Coeffcient Estimates CI (95%) p-Value
Intercept 3.50 3.08 – 3.93 <0.001
SES1 0.02 -0.01 – 0.04 0.138
TBV -0.00 -0.00 – 0.00 0.187
Age -0.00 -0.01 – 0.00 0.440
Sex 0.08 -0.02 – 0.17 0.116
Observations 69
R2 / R2 adjusted 0.100 / 0.044
  rh_middletemporal_CT
Coeffcient Estimates CI (95%) p-Value
Intercept 3.10 2.76 – 3.44 <0.001
SES1 0.02 -0.00 – 0.03 0.064
TBV -0.00 -0.00 – 0.00 0.968
Age -0.00 -0.01 – -0.00 0.006
Sex 0.02 -0.06 – 0.09 0.667
Observations 69
R2 / R2 adjusted 0.185 / 0.134

CT Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models: MEL

MEL Results Q>=4

  • MEL not a significant predictor of cortical thickness in language regions, controlling for age and sex.

MEL Results Q>=3

  • MEL not a significant predictor of cortical thickness in language regions, controlling for age and sex.

CT Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models: ZipIncome

Zip-Income Results, Q>=4

  • Zip-Income not a significant predictor of CT in language regions, controlling for age in sex.

Zip-Income Results, Q>=3

  • Zip-Income not a significant predictor of CT in language regions, controlling for age in sex.

CT Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models: Dx

Dx Results, Q>=4

  lh_parsorbitalis_CT
Coeffcient Estimates CI (95%) p-Value
Intercept 3.54 3.31 – 3.78 <0.001
Dx 0.10 -0.01 – 0.21 0.069
Age -0.01 -0.01 – -0.01 <0.001
Observations 55
R2 / R2 adjusted 0.337 / 0.311

  • Dx predicts left hemisphere pars orbitalis CT, controlling for age (sex not a significant predictor) in the Q>=4 dataset.

Dx Results, Q>=3

  lh_parsorbitalis_CT
Coeffcient Estimates CI (95%) p-Value
Intercept 3.42 3.17 – 3.67 <0.001
Dx 0.15 0.03 – 0.27 0.014
Age -0.01 -0.01 – -0.01 <0.001
Observations 70
R2 / R2 adjusted 0.288 / 0.267

  • Dx predicts left hemisphere pars orbitalis CT, controlling for age (sex not a significant predictor) in the Q>=3 dataset.

CT Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models: Sex/Gender

Sex Results, Q>=4

  • Sex/gender predicts right hemisphere pars triangularis CT, controlling for age and dx.
  rh_parstriangularis_CT
Coeffcient Estimates CI (95%) p-Value
Intercept 2.93 2.70 – 3.16 <0.001
Sex 0.08 -0.00 – 0.16 0.059
Age -0.01 -0.01 – -0.00 <0.001
Dx 0.04 -0.04 – 0.12 0.373
Observations 55
R2 / R2 adjusted 0.268 / 0.225

  • Sex predicts right hemisphere pars orbitalis CT, controlling for age and dx in the Q>=4 dataset.

Sex/Gender Results, Q>=3

  • Sex predicts right hemisphere pars orbitalis CT, controlling for age and dx in the Q>=3 dataset.
  lh_parsorbitalis_CT
Coeffcient Estimates CI (95%) p-Value
Intercept 2.76 2.52 – 2.99 <0.001
Sex 0.09 0.00 – 0.18 0.050
Age -0.00 -0.01 – -0.00 0.041
Dx 0.05 -0.03 – 0.14 0.228
Observations 70
R2 / R2 adjusted 0.115 / 0.074

SA Results, Bivariate Correlations

Bivariate Correlation Plots, ASD&TD Together, Q>=4

Bivariate Plot

Bivariate Plot

Bivariate Correlation ASD and TD, Q>=3

Bivariate Plot

Bivariate Plot

Bivariate correlations, ASD Only, Q>=4

Bivariate Plot ASD Only

Bivariate Plot ASD Only

Bivariate correlations, ASD Only, Q>=3

Bivariate Plot ASD Only

Bivariate Plot ASD Only

Bivariate correlations, TD Only Q>=4

Bivariate Plot, TD Only

Bivariate Plot, TD Only

Bivariate correlations, TD Only Q>=3

Bivariate Plot, TD Only

Bivariate Plot, TD Only

SA Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models: INR

INR Results, Q>=4

  • INR Predicts rh pars orbitalis surface area controlling for TBV, age, and sex (although neither age nor sex were significant predictors of RH pars orbitalis SA). Dx is also not a significant predictor, and there is no significant Dx by INR interaction.
  rh_parsorbitalis_area
Coeffcient Estimates CI (95%) p-Value
Intercept 16.67 -267.00 – 300.34 0.906
INR -9.01 -19.57 – 1.55 0.093
TBV 0.00 0.00 – 0.00 <0.001
Age 1.37 -0.94 – 3.69 0.238
Sex -24.58 -88.26 – 39.10 0.442
Observations 54
R2 / R2 adjusted 0.485 / 0.443

INR Results, Q>=3

  • INR Predicts rh pars orbitalis surface area controlling for TBV, age, and sex (although neither age nor sex were significant predictors of RH pars orbitalis SA).
  rh_parsorbitalis_area
Coeffcient Estimates CI (95%) p-Value
Intercept -18.25 -242.38 – 205.87 0.871
INR -9.58 -18.83 – -0.34 0.042
TBV 0.00 0.00 – 0.00 <0.001
Age 0.78 -1.23 – 2.79 0.439
Sex -35.71 -86.48 – 15.06 0.165
Observations 68
R2 / R2 adjusted 0.522 / 0.492

SA Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models: SES1

SES1 Results, Q>=4

  • Neighborhood advantage is associated with left hemisphere pars orbitalis and pars opercularis SA, controlling for TBV and age.
  lh_parsorbitalis_area
Coeffcient Estimates CI (95%) p-Value
Intercept -52.58 -238.51 – 133.36 0.573
SES1 -9.95 -18.83 – -1.06 0.029
TBV 0.00 0.00 – 0.00 <0.001
Age 2.39 0.90 – 3.87 0.002
Observations 55
R2 / R2 adjusted 0.611 / 0.588

  lh_parsopercularis_area
Coeffcient Estimates CI (95%) p-Value
Intercept -79.08 -640.81 – 482.65 0.779
SES1 -28.64 -55.47 – -1.80 0.037
TBV 0.00 0.00 – 0.00 <0.001
Age 6.39 1.91 – 10.86 0.006
Observations 55
R2 / R2 adjusted 0.541 / 0.514

SES1 Results, Q>=3

  • Neighborhood advantage associated with left hemisphere and pars orbitalis SA, as in the Q>=4 dataset, not significantly associated with pars opercularis SA in this dataset.

  • Neighborhood advantage is also associated with *as well as right hemisphere pars triangularis SA (not significant in the Q>=4 dataset

  lh_parsorbitalis_area
Coeffcient Estimates CI (95%) p-Value
Intercept -57.59 -225.59 – 110.41 0.496
SES1 -9.84 -18.66 – -1.02 0.029
TBV 0.00 0.00 – 0.00 <0.001
Age 2.36 0.94 – 3.79 0.002
Observations 69
R2 / R2 adjusted 0.593 / 0.574

  lh_parsopercularis_area
Coeffcient Estimates CI (95%) p-Value
Intercept 123.92 -383.01 – 630.85 0.627
SES1 -17.55 -44.16 – 9.06 0.192
TBV 0.00 0.00 – 0.00 <0.001
Age 5.60 1.30 – 9.91 0.012
Observations 69
R2 / R2 adjusted 0.455 / 0.430
  rh_parstriangularis_area
Coeffcient Estimates CI (95%) p-Value
Intercept -470.22 -1057.97 – 117.54 0.115
SES1 -38.02 -68.88 – -7.17 0.017
TBV 0.00 0.00 – 0.00 <0.001
Age 4.08 -0.91 – 9.08 0.108
Observations 69
R2 / R2 adjusted 0.485 / 0.461

SA Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models: MEL

MEL Results, Q>=4

  • Maternal educational level is not associated with surface area in language regions.

MEL Results, Q>=3

  • Maternal educational level is not associated with surface area in language regions.

SA Results, compare results for data quality Q>=4 vs. data quality Q>=3, Regression Models: Zip-Income

Zip-Income Results, Q>=4

  • Zip-Income is negatively associated with lh pars orbitalis SA in the Q>=4 dataset.
  lh_parsorbitalis_area
Coeffcient Estimates CI (95%) p-Value
Intercept -0.28 -212.60 – 212.04 0.998
Zip-Income -0.00 -0.00 – -0.00 0.039
TBV 0.00 0.00 – 0.00 <0.001
Age 1.52 -0.13 – 3.17 0.069
Observations 44
R2 / R2 adjusted 0.562 / 0.530

Zip-Income Results, Q>=3

  • Zip-Income is not associated with rh pars triangularis SA in the Q>=3 dataset.
  rh_parstriangularis_area
Coeffcient Estimates CI (95%) p-Value
Intercept 159.08 -273.97 – 592.13 0.465
Zip-Income -0.00 -0.00 – 0.00 0.777
TBV 0.00 0.00 – 0.00 <0.001
Age 3.74 -0.04 – 7.51 0.052
Observations 58
R2 / R2 adjusted 0.448 / 0.418